In a bold move that has sent shockwaves through the AI industry, Chinese tech giant Alibaba (9988.HK) has unveiled Qwen 2.5-Max, a new iteration of its artificial intelligence model. The timing of the release—coinciding with the first day of the Lunar New Year, when most of China is on holiday—shows the mounting pressure from the meteoric rise of AI startup DeepSeek.
Alibaba’s cloud unit announced on WeChat that Qwen 2.5-Max outperforms OpenAI’s GPT-4o, DeepSeek-V3, and Meta’s Llama-3.1-405B across multiple benchmarks, reinforcing its position as a major player in the AI arms race. The rapid rise of DeepSeek over the past three weeks has not only rattled global AI competitors but also put immense pressure on domestic tech giants to step up their game.
DeepSeek has been making waves since the launch of its DeepSeek-V3-powered AI assistant on January 10, followed by the release of its R1 model on January 20. The Chinese startup’s ability to develop cutting-edge AI at a fraction of the cost has shocked Silicon Valley, causing tech stocks to tumble and raising concerns about the sustainability of massive AI investments in the West.
The disruptive success of DeepSeek has triggered an urgent response from China’s biggest tech firms. Just two days after DeepSeek-R1 debuted, ByteDance, the parent company of TikTok, rolled out an update to its flagship AI model, boasting that it surpassed Microsoft-backed OpenAI’s o1 in the AIME benchmark test. DeepSeek had similarly claimed that its R1 model rivalled OpenAI’s o1 in various performance metrics.
The AI battle in China has been heating up since last May when DeepSeek-V2 sparked a price war by offering its open-source model at an unprecedentedly low cost—just 1 yuan ($0.14) per 1 million tokens. In response, Alibaba slashed prices on its AI models by up to 97%, forcing other major players like Baidu (9888.HK) and Tencent (0700.HK) to follow suit.
Read also: AI WAR: OpenAI accuses China’s DeepSeek of stealing ChatGPT secrets
DeepSeek’s enigmatic founder, Liang Wenfeng, has remained unfazed by the price wars, stating in a rare interview that the company’s primary goal is to achieve artificial general intelligence (AGI), a breakthrough that could revolutionize AI capabilities worldwide. OpenAI defines AGI as systems capable of surpassing human performance in most economically valuable tasks.
Unlike tech behemoths such as Alibaba, which employ hundreds of thousands of workers, DeepSeek operates more like a research lab, staffed mainly by young PhD graduates from China’s top universities. Liang has openly criticized the rigid structures of China’s largest tech firms, suggesting that their bureaucratic approach and high costs may hinder their ability to innovate in the AI space.
With the race for AI dominance intensifying, Alibaba’s Qwen 2.5-Max enters the battlefield at a critical juncture. Whether it can truly outperform DeepSeek and global competitors remains to be seen, but one thing is clear: China’s AI war is just getting started, and the stakes have never been higher.
